The following description provides a summary of information relevant to the present invention. It is not an admission that any of the information provided herein is prior art to the presently claimed invention, nor that any of the publications specifically or implicitly referenced are prior art to that invention.
Glutamate is the major excitatory neurotransmitter in the mammalian brain. Glutamate produces its effects on central neurons by binding to and thereby activating cell surface receptors. These receptors have been subdivided into two major classes, the ionotropic and metabotropic glutamate receptors, based on the structural features of the receptor proteins, the means by which the receptors transduce signals into the cell, and pharmacological profiles.
The ionotropic glutamate receptors (iGluRs) are ligand-gated ion channels that, upon binding glutamate, open to allow the selective influx of certain monovalent and divalent cations, thereby depolarizing the cell membrane. In addition, certain iGluRs with relatively high calcium permeability can activate a variety of calcium-dependent intracellular processes. These receptors are multisubunit protein complexes that may be homomeric or heteromeric in nature. The various iGluR subunits all share common structural motifs, including a relatively large amino-terminal extracellular domain (ECD), followed by a multiple transmembrane domain (TMD) comprising two membrane-spanning regions (TMs), a second smaller intracellular loop, and a third TM, before terminating with an intracellular carboxy-terminal domain (CT). Historically the iGluRs were first subdivided pharmacologically into three classes based on preferential activation by the agonists alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionic acid (AMPA), kainate (KA), and N-methyl-D-aspartate (NMDA). Later, molecular cloning studies coupled with additional pharmacological studies revealed a greater diversity of iGluRs, in that multiple subtypes of AMPA, KA and NMDA receptors are expressed in the mammalian CNS (Hollman and Heinemann, Ann. Rev. Neurosci. 7:31, 1994).
The metabotropic glutamate receptors (mGluRs) are G protein-coupled receptors capable of activating a variety of intracellular second messenger systems following the binding of glutamate or other potent agonists including quisqualate and 1-aminocyclopentane-1,3-dicarboxylic acid (trans-ACPD) (Schoepp et al., Trends Pharmacol. Sci. 11:508, 1990; Schoepp and Conn, Trends Pharmacol. Sci. 14:13, 1993).
Activation of different metabotropic glutamate receptor subtypes in situ elicits one or more of the following responses: activation of phospholipase C, increases in phosphoinositide (PI) hydrolysis, intracellular calcium release, activation of phospholipase D, activation or inhibition of adenylyl cyclase, increases and decreases in the formation of cyclic adenosine monophosphate (cAMP), activation of guanylyl cyclase, increases in the formation of cyclic guanosine monophosphate (cGMP), activation of phospholipase A.sub.2, increases in arachidonic acid release, and increases or decreases in the activity of voltage- and ligand-gated ion channels (Schoepp and Conn, Trends Pharmacol. Sci. 14:13, 1993; Schoepp, Neurochem. Int. 24:439, 1994; Pin and Duvoisin, Neuropharmacology 34:1, 1995).
Thus far, eight distinct mGluR subtypes have been isolated via molecular cloning, and named mGluR1 to mGluR8 according to the order in which they were discovered (Nakanishi, Neuron 13:1031, 1994, Pin and Duvoisin, Neuropharmacology 34:1, 1995; Knopfel et al., J. Med. Chem. 38:1417, 1995). Further diversity occurs through the expression of alternatively spliced forms of certain mGluR subtypes (Pin et al., PNAS 89:10331, 1992; Minakami et al., BBRC 199:1136, 1994). All of the mGluRs are structurally similar, in that they are single subunit membrane proteins possessing a large amino-terminal extracellular domain (ECD) followed by seven putative transmembrane domain (7TMD) comprising seven putative membrane spanning helices connected by three intracellular and three extracellular loops, and an intracellular carboxy-terminal domain of variable length (cytoplasmic tail) (CT) (see, Schematic FIG. 1a).
The eight mGluRs have been subdivided into three groups based on amino acid sequence identities, the second messenger systems they utilize, and pharmacological characteristics (Nakanishi, Neuron 13:1031, 1994; Pine and Duvoisin, Neuropharmacology 34:1, 1995; Knopfel et al., J. Med. Chem. 38:1417, 1995). The amino acid identity between mGluRs within a given group is approximately 70% but drops to about 40% between mGluRs in different groups. For mGluRs in the same group, this relatedness is roughly paralleled by similarities in signal transduction mechanisms and pharmacological characteristics.
The Group I mGluRs comprise mGluR1, mGluR5 and their alternatively spliced variants. The binding of agonists to these receptors results in the activation of phospholipase C and the subsequent mobilization of intracellular calcium. For example, Xenopus oocytes expressing recombinant mGluR1 receptors have been utilized to demonstrate this effect indirectly by electrophysiological means (Masu et al., Nature 349:760, 1991; Pin et al., PNAS 89:10331, 1992). Similar results were achieved with oocytes expressing recombinant mGluR5 receptors (Abe et al., J. Biol. Chem. 267:13361, 1992; Minakami et al., BBRC 199:1136, 1994). Alternatively, agonist activation of recombinant mGluR1 receptors expressed in Chinese hamster ovary (CHO) cells stimulated PI hydrolysis, cAMP formation, and arachidonic acid release as measured by standard biochemical assays (Aramori and Nakanishi, Neuron 8:757, 1992). In comparison, activation of mGluR5 receptors expressed in CHO cells stimulated PI hydrolysis and subsequent intracellular calcium transients but no stimulation of cAMP formation or arachidonic acid release was observed (Abe et al., J. Biol. Chem. 267:13361, 1992). The agonist potency profile for Group I mGluRs is quisqualate&gt;glutamate=ibotenate&gt;(2S,1'S,2'S) -2-carboxycyclopropyl)glycine (L-CCG-I)&gt;(1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (ACPD). Quisqualate is relatively selective for Group I receptors, as compared to Group II and Group III mGluRs, but it also potently activates ionotropic AMPA receptors (Pin and Duvoisin, Neuropharmacology, 34:1, Knopfel et al., J. Med. Chem. 38:1417, 1995).
The Group II mGluRs include mGluR2 and mGluR3. Activation of these receptors as expressed in CHO cells inhibits adenylyl cyclase activity via the inhibitory G protein, G.sub.i, in a pertussis toxin-sensitive fashion (Tanabe et al., Neuron 8:169, 1992; Tanabe et al., Neurosci. 13:1372, 1993). The agonist potency profile for Group II receptors is L-CCG-I&gt;glutamate&gt;ACPD&gt;ibotenate&gt;quisqualate. Preliminary studies suggest that L-CCG-I and (2S,1'R,2'R,3'R)-2-(2,3-dicarboxycyclopropyl)glycine (DCG-IV) are both relatively selective agonists for the Group II receptors (Knopfel et al., J. Med. Chem. 38:1417, 1995).
The Group III mGluRs include mGluR4, mGluR6, mGluR7 and mGluR8. Like the Group II receptors these mGluRs are negatively coupled to adenylate cyclase to inhibit intracellular cAMP accumulation in a pertussis toxin-sensitive fashion when expressed in CHO cells (Tanabe et al., J. Neurosci. 13:1372, 1993; Nakajima et al., J. Biol. Chem. 268:11868, 1993; Okamoto et al., J. Biol. Chem. 269:1231, 1994; Duvoisin et al., J. Neurosci. 15:3075, 1995). As a group, their agonist potency profile is (S)-2-amino-4-phosphonobutyric acid (L-AP4)&gt;glutamate&gt;ACPD&gt;quisqualate, but mGluR8 may differ slightly with glutamate being more potent than L-AP4 (Knopfel et al., J. Med. Chem. 38:1417, 1995; Duvoisin et al., J. Neurosci. 15: 3075, 1995). Both L-AP4 and (S)-serine-O-phosphate (L-SOP) are relatively selective agonists for the Group III receptors.
Finally, the eight mGluR subtypes have unique patterns of expression within the mammalian CNS that in many instances are overlapping (Masu et al., Nature 349:760, 1991; Martin et al., Neuron 9:259, 1992; Ohishi et al., Neurosci. 53:1009, 1993; Tanabe et al., J. Neurosci. 13:1372; Ohishi et al., Neuron 13:55, 1994, Abe et al., J. Biol. Chem. 267:13361, 1992; Nakajima et al., J. Biol. Chem. 268:11868, 1993; Okamoto et al., J. Biol. Chem. 269:1231, 1994; Duvoisin et al., J. Neurosci. 15:3075, 1995). As a result certain neurons may express only one particular mGluR subtype, while other neurons may express multiple subtypes that may be localized to similar and/or different locations on the cell (i.e., postsynaptic dendrites and/or cell bodies versus presynaptic axon terminals). Therefore, the functional consequences of mGluR activation on a given neuron will depend on the particular mGluRs being expressed; the receptors' affinities for glutamate and the concentrations of glutamate the cell is exposed to; the signal transduction pathways activated by the receptors; and the locations of the receptors on the cell. A further level of complexity may be introduced by multiple interactions between mGluR expressing neurons in a given brain region. As a result of these complexities, and the lack of subtype-specific mGluR agonists and antagonists, the roles of particular mGluRs in physiological and pathophysiological processes affecting neuronal function are not well defined. Still, work with the available agonists and antagonists have yielded some general insights about the Group I mGluRs as compared to the Group II and Group III mGluRs.
Attempts at elucidating the physiological roles of Group I mGluRs suggest that activation of these receptors elicits neuronal excitation. Various studies have demonstrated that ACPD can produce postsynaptic excitation upon application to neurons in the hippocampus, cerebral cortex, cerebellum, and thalamus as well as other brain regions. Evidence indicates that this excitation is due to direct activation of postsynaptic mGluRs, but it has also been suggested to be mediated by activation of presynaptic mGluRs resulting in increased neurotransmitter release (Baskys, Trends Pharmacol. Sci. 15:92, 1992; Schoepp, Neurochem. Int. 24:439, 1994; Pin and Duvoisin, Neuropharmacology 34:1). Pharmacological experiments implicate Group I mGluRs as the mediators of this excitation. The effect of ACPD can be reproduced by low concentrations of quisqualate in the presence of iGluR antagonists (Hu and Storm, Brain Res. 568:339, 1991; Greene et al. Eur. J. Pharmacol. 226:279, 1992), and two phenylglycine compounds known to activate mGluR1, (S)-3-hydroxyphenylglycine ((S)-3HPG) and (S)-3,5-dihydroxyphenylglycine ((S)-DHPG), also produce the excitation (Watkins and Collingridge, Trends Pharmacol. Sci. 15:333, 1994). In addition, the excitation can be blocked by (S)-4-carboxyphenylglycine ((S)-4CPG), (S)-4-carboxy-3-hydroxyphenylglycine ((S)-4C3HPG) and (+)-alphamethyl-4-carboxyphenylglycine ((+)-MCPG), compounds known to be mGluR1 antagonists (Eaton et al., Eur. J. Pharmacol. 244:195, 1993; Watkins and Collingridge, Trends Pharmacol. Sci. 15:333, 1994).
Other studies examining the physiological roles of mGluRs indicate that activation of presynaptic mGluRs can block both excitatory and inhibitory synaptic transmission by inhibiting neurotransmitter release (Pin and Duvoisin, Neuropharmacology 34:1). Presynaptic blockade of excitatory synaptic transmission by ACPD has been observed on neurons in the visual cortex, cerebellum, hippocampus, striatum and amygdala (Pin et al., Curr. Drugs: Neurodegenerative Disorders 1:111, 1993), while similar blockade of inhibitory synaptic transmission has been demonstrated in the striatum and olfactory bulb (Calabresi et al., Neurosci. Lett. 139:41, 1992; Hayashi et al., Nature 366:687, 1993). Multiple pieces of evidence suggest that Group II mGluRs mediate this presynaptic inhibition. Group II mGluRs are strongly coupled to inhibition of adenylyl cyclase, like alpha.sub.2 -adrenergic and SHT.sub.1A -serotonergic receptors which are known to mediate presynaptic inhibition of neurotransmitter release in other neurons. The inhibitory effects of ACPD can also be mimicked by L-CCG-I and DCG-IV, which are selective agonists at Group II mGluRs (Hayashi et al., Nature 366:687, 1993; Jane et al., Br. J. Pharmacol. 112:809, 1994). Moreover, it has been demonstrated that activation of mGluR2 can strongly inhibit presynaptic, N-type calcium channel activity when the receptor is expressed in sympathetic neurons (Ikeda et al., Neuron 14:1029, 1995), and inactivation of these channels is known to inhibit neurotransmitter release. Finally, it has been observed that L-CCG-I, at concentrations selective for Group II mGluRs, inhibits the depolarization-evoked release of .sup.3 H-aspartate from rat striatal slices (Lombardi et al., Br. J. Pharmacol. 110:1407, 1993). Evidence for physiological effects of Group II mGluR activation at the postsynaptic level is limited. However, one study suggests that postsynaptic actions of L-CCG-I can inhibit NMDA receptor activation in cultured mesencephalic neurons (Ambrosini et al., Mol. Pharmacol. 47:1057, 1995).
Physiological studies have demonstrated that L-AP4 can also inhibit excitatory synaptic transmission on a variety of CNS neurons. Included are neurons in the cortex, hippocampus, amygdala, olfactory bulb and spinal cord (Koerner and Johnson, Excitatory Amino Acid Receptors; Design of Agonists and Antagonists p. 308, 1992; Pin et al., Curr. Drugs: Neurodegenerative Disorders 1:111, 1993). The accumulated evidence indicates that the inhibition is mediated by activation of presynaptic mGluRs. Since the effects of L-AP4 can be mimicked by L-SOP, and these two agonists are selective for Group III mGluRs, members of this mGluR group are implicated as the mediators of the presynaptic inhibition (Schoepp, Neurochem. Int. 24:439, 1994; Pin and Duvoisin, Neuropharmacology 34:1). In olfactory bulb neurons it has been demonstrated that L-AP4 activation of mGluRs inhibits presynaptic calcium currents (Trombley and Westbrook, J. Neurosci. 12:2043, 1992). It is therefore likely that the mechanism of presynaptic inhibition produced by activation of Group III mGluRs is similar to that for Group II mGluRs, i.e., blockade of N-type calcium channels and inhibition of neurotransmitter release. L-AP4 is also known to act postsynaptically to hyperpolarize ON bipolar cells in the retina. It has been suggested that this action may be due to activation of a mGluR, which is coupled to the cGMP phosphodiesterase in these cells (Schoepp, Neurochem. Int. 24:439, 1994; Pin and Duvoisin, Neuropharmacology 34:1).
Metabotropic glutamate receptor activation studies using agonists, antagonists and recombinant vertebrate cell lines expressing mGluRs have been used to evaluate the cellular effects of the stimulation and the inhibition of different metabotropic glutamate receptors. For example, agonist stimulation of mGluR1 expressed in Xenopus oocytes demonstrated coupling of receptor activation to mobilization of intracellular calcium as assessed indirectly using electrophysiology techniques (Masu et al., Nature 349:760-765, 1991). Agonist stimulation of mGluR1 expressed in CHO cells stimulated PI hydrolysis, cAMP formation and arachidonic acid release (Aramori and Nakanishi, Neuron 8:757-765, 1992). Agonist stimulation of mGluR5 expressed in CHO cells also stimulated PI hydrolysis which was shown to be associated with a transient increase in cytosolic calcium as assessed by loading cells with the fluorescent calcium chelator fura-2 (Abe et al., J. Biol. Chem. 267:13361-13368, 1992). Agonist-induced activation of mGluR1 and mGluR5 induced PI hydrolysis in CHO cells was not antagonized by AP3 and AP4, which are both antagonists of glutamate-stimulated PI hydrolysis in situ (Nicoletti et al., Proc. Natl. Acad. Sci. USA 833:1931-1935, 1986; Schoepp and Johnson, J. Neurochem. 53:273-278, 1989). Agonist stimulation of CHO cells expressing mGluR2 (Tanabe et al., Neuron 8:169-179, 1992) or mGluR7 (Okamoto et al., J. Biol. Chem. 269:1231-1236, 1994) resulted in receptor-mediated inhibition of cAMP formation and also confirmed the ligand specificity previously observed in situ. Studies using agonists were also carried out in conjunction with site-directed mutagenesis to reveal specific amino acids playing important roles in glutamate binding (O'Hara et al., Neuron 11:41-52, 1993).
Metabotropic glutamate receptors (mGluRs) have been implicated in a variety of neurological pathologies including stroke, head trauma, spinal cord injury, epilepsy, ischemia, hypoglycemia, anoxia, and neurodegenerative diseases such as Alzheimer's disease (Schoepp and Conn, Trends Pharmacol. Sci. 14:13, 1993; Cunningham et al., Life Sci. 54: 135, 1994; Pin et al., Neuropharmacology 34:1, 1995; Knopfel et al., J. Med. Chem. 38:1417, 1995;). A role for metabotropic glutamate receptors in nociception and analgesia has also been demonstrated (Meller et al., Neuroreport 4:879, 1993). Metabotropic glutamate receptors have also been shown to be required for the induction of hippocampal long-term potentiation and cerebellar long-term depression (Bashir et al., Nature 363:347, 1993; Bortolotto et al., Nature 368:740, 1994; Aiba et. al. Cell 79: 365 and Cell 79: 377, 1994).
Metabotropic glutamate receptor agonists have been reported to have effects on various physiological activities. For example, trans-ACPD was reported to possess both proconvulsant and anticonvulsant effects (Zheng and Gallagher, Neurosci. Lett. 125:147, 1991; Sacaan and Schoepp, Neurosci. Lett. 139:77, 1992; Taschenberger et al., Neuroreport 3:629, 1992; Sheardown, Neuroreport 3:916, 1992), and neuroprotective effects in vitro and in vivo (Pizzi et al., J. Neurochem. 61:683, 1993; Koh et al., Proc. Natl. Acad. Sci. USA 88:9431, 1991; Birrell et al., Neuropharmacol. 32:1351, 1993; Siliprandi et al., Eur. J. Pharmacol. 219:173, 1992; Chiamulera et al., Eur. J. Pharmacol. 216:335, 1992). The metabotropic glutamate receptor antagonist L-AP3 was shown to protect against hypoxic injury in vitro (Opitz and Reymann, Neuroreport 2:455, 1991). A subsequent study reported that trans-ACPD produced neuroprotection which was antagonized by L-AP3 (Opitz and Reymann, Neuropharmacol. 32:103, 1993). (5)-4C3HPG was shown to protect against audiogenic seizures in DBA/2 mice (Thomasen et al., J. Neurochem. 62:2492, 1994). Other modulatory effects expected of metabotropic glutamate receptor modulators include synaptic transmission, neuronal death, neuronal development, synaptic plasticity, spatial learning, olfactory memory, central control of cardiac activity, waking, control of movements, and control of vestibule ocular reflex (for reviews, see Nakanishi, Neuron 13:1031-37, 1994; Pin et al., Neuropharmacology 34:1, 1995; Knopfel et al., J. Med. Chem. 38:1417, 1995).
The structures of mGluR-active molecules currently known in the art are limited to amino acids which appear to act by binding at the glutamate binding site (Pin, et al, Neuropharmacology 34:1, 1995; Knopfel et al., J. Med. Chem. 38:1418). This limits the range of pharmacological properties and potential therapeutic utilities of such compounds. Furthermore, the range of pharmacological specificities associated with these mGluR-active molecules does not allow for complete discrimination between different subtypes of metabotropic glutamate receptors (Pin et al., Neuropharmacology 34:1, 1995 and Knopfel et al., J. Med. Chem. 38:1418). Rapid progress in the field of mGluR-active molecules cannot be made until more potent and more selective mGluR agonists, antagonists and modulators are discovered (Pin et al., Neuropharmacology 34:1, 1995; Knopfel et al., J. Med. Chem. 38:1418). Indeed, no mGluR-active molecules are presently under clinical development. High throughput functional screening of compounds and compound libraries using cell lines expressing individual mGluRs represents an important approach to identifying such novel compounds (Knopfel et al., J. Med. Chem. 38:1418).
Several laboratories have constructed cell lines expressing metabotropic glutamate receptors which appear to function appropriately (Abe et al., J. Biol. Chem. 267:13361, 1992; Tanabe et al., Neuron 8:169, 1992; Aramori and Nakanishi, Neuron 8:757, 1992, Nakanishi, Science 258:597, 1992; Thomsen et al., Brain Res. 619:22, 1992; Thomsen et al., Eur. J. Pharmacol. 227:361, 1992; O'Hara et al., Neuron 11:41, 1993; Nakjima et al., J. Biol. Chem. 268:11868, 1993; Tanabe et al., J. Neurosci. 13:1372, 1993; Saugstad et al., Mol. Pharmacol. 45:367, 1994; Okamoto et al., J. Biol. Chem. 269:1231, 1994; Gabellini et al., Neurochem. Int. 24:533, 1994; Lin et al., Soc. Neurosci. Abstr. 20:468, 1994; Flor et al., Soc. Neurosci. Abstr. 20:468, 1994; Flor et al., Neuropharmacology 34:149, 1994). Other reports have noted that expression of functional mGluR expressing cell lines is not predictable. For example, Tanabe et al., (Neuron 8:169, 1992) were unable to demonstrate functional expression of mGluR3 and mGluR4, and noted difficulty obtaining expression of native mGluR1 in CHO cells. Gabellini et al., (Neurochem. Int. 24:533, 1994) also noted difficulties with mGluR1 expression in HEK 293 cells and it is possible that some of these difficulties may be due to desensitization characteristics of these receptors. Furthermore, screening methodologies useful for identification of compounds active at Class I mGluRs are not readily amenable to identification of compounds active at class II and III mGluRs and vice versa due to the differences in second messenger coupling. Finally, mGluRs have been noted to rapidly desensitize upon agonist stimulation which may adversely affect the viability of cell lines expressing these receptors and makes the use of native mGluRs for screening difficult.
Different G-protein coupled receptors exhibit differential ligand affinities and coupling to second messengers. G-protein coupled receptors all have a similar structure: an N-terminal extracellular domain (ECD), a seven-transmembrane domain (7TMD) comprising seven membrane spanning helices and therefore defining three intracellular and three extracellular loops, and a cytoplasmic tail (CT), but differ in the exact sequences comprising each region. These sequence differences are thought to provide the specificity of receptor interactions with ligands of different chemical compositions and receptor interaction with different G-proteins. Construction of chimeric receptors in which small peptide segments from related receptors are exchanged using recombinant DNA techniques has proven a useful technique to assess the participation of different sequence regions in determining this specificity. For example, exchanging the third intracellular loops between various adrenergic, muscarinic acetylcholine and angiotensin receptors results in conversion of G-protein coupling specificity. Thus, receptors whose activation normally results in inhibition or activation of adenylate cyclase can be converted to receptors with the same or similar ligand binding properties but whose activation leads to stimulation of phospholipase-C and vice versa (Kobilka et al., Science 240:1310, 1988; Wess et al., FEBS Lett. 258:133, 1989; Cotecchia et al., Proc. Nat'l. Acad. Sci. U.S.A. 87:2896, 1990; Lechleiter et al., EMBO J. 9:4381, 1990; Wess et al., Mol. Pharmacol. 38:517, 1990; Wong et al., J. Biol. Chem. 265:6219, 1990; Cotecchia et al., J. Biol. Chem. 267:1633, 1992; Wang et al., J. Biol. Chem. 270:16677, 1995). In these receptors which share the third intracellular loop plays an important role in determining the specificity of G-protein coupling. While such experiments indicate that the third intracellular loop plays an important role in determining the specificity of G protein coupling in these related receptors, they have failed to identify any specific amino acid sequence motif which is responsible. In addition, the third intracellular loop has been shown to be at least partly responsible for desensitization of such receptors (Okamoto et al., Cell 67:723, 1991; Liggett et al., J. Biol. Chem. 267:4740, 1992).
Metabotropic glutamate receptors are related to other G-protein coupled receptors in overall topology, but not in specific amino acid sequence. An unusual feature of mGluRs is their very large ECDs (ca. 600 amino acids). In many other G-protein coupled receptors, ligand binding takes place within the 7TMD. However, the large ECD of each mGluR is thought to provide the ligand binding determinants (Nakanishi, Science 258:597, 1992; O'Hara et al., Neuron 11:41, 1993; Shigemoto et al., Neuron 12:1245, 1994). Chimeric mGluRs in which the ECDs of mGluRs with different ligand affinities and different G-protein coupling are exchanged have been used to demonstrate that the ECD of mGluRs defines ligand specificity but not G-protein specificity (Takahashi et al., J. Bio. Chem. 268:19341, 1993). Also unlike other G-protein coupled receptors in which the third intracellular loop is variable in size and sequence, the third intracellular loops of mGluRs are small and extremely well conserved (Brown E. M. et al., Nature 366:575, 1993). Chimeric mGluRs have been prepared in which the second intracellular loops and/or cytoplasmic tails were exchanged (Pin et al., EMBO J. 13:342). These experiments lead the investigators to conclude that unlike most other G-protein coupled receptors, "both the C-terminal end of the second intracellular loop and the segment located downstream of the seventh transmembrane domain are necessary for the specific activation of phospholipase-C by mGluR1c" and to suggest that the second intracellular loop of mGluRs plays the role of the third intracellular loop of other G-protein coupled receptors.
Naturally occurring mRNA splice variants have been noted to produce prostaglandin E3 (EP3) receptors with essentially identical ligand binding properties but which preferentially activate different second messenger pathways (differential G-protein coupling) and which exhibit different desensitization properties (Namba et al., Nature 365:166, 1993; Sgimoto et al., J. Biol. Chem. 268:2712, 1993; Negishi et al., J. Biol. Chem. 268:9517, 1993). These variant receptor isoforms differ only in their cytoplasmic tails. The isoforms with the longer tails couple efficiently to phospholipase-C while those with the shorter tails do not. However, analyses of naturally occurring mRNA splice variants of mGluR1 and mGluR5 have indicated that their long cytoplasmic tails may not be directly involved in G protein coupling (Pin et al., Proc. Nat'l. Acad. Sci. U.S.A. 89:10331, 1992; Joly et al., J. Neuroscience 15:3970, 1995). In fact, Pin et al., (supra) have stated that "The very long C-terminal domain found only in PLC-coupled mGluRs (mGluR1 and 5) is, however, probably not involved in the specific interaction with PLC-activating G proteins."
Recently, calcium receptor has been described (Brown E. M. et al., Nature 366:575, 1993; Riccardi D., et al., Proc. Nat'l. Acad. Sci. USA 92:131-135, 1995; Garrett J. E., et al., J. Biol. Chem. 31:12919-12925, 1995). This CaR is the only known receptor which exhibits significant sequence homology with mGluRs except for other mGluRs. The CaR exhibits about .about.25% sequence homology (amino acid identities) to any one mGluR while mGluRs are &gt;40% homologous (amino acid identities) to one another. The CaR is structurally related to mGluRs having a large ECD which has been implicated in receptor function and probable ligand binding (Brown E. M. et al., Nature 366:575, 1993; Pollak, M. R., et al., Cell 75:1297-1303, 1993). This similarity of structure does not confer close similarity in ligand binding specificity since the native ligand for the CaR is the inorganic ion, Ca.sup.2+, and glutamate does not modulate CaR activity. The CaR also has a large cytoplasmic tail and is coupled to the stimulation of phospholipase-C. Thus, the CaR is structurally and functionally more related to mGluR1 and 5 than to other mGluRs. Pin et al., (EMBO J. 13:342, 1994) have noted that certain amino acids are conserved within the intracellular loops of mGluRs which couple to phospholipase-C and different amino acids are conserved in these same positions within the intracellular loops of mGluRs which couple to the inhibition of adenylate cyclase. Intracellular loops 1 and 3 are the most highly conserved sequences between mGluRs and the CaR (Brown E. M. et al., Nature 366:575, 1993), but only about half of these particular amino acids are found in the corresponding position of the CaR and only one of these is actually the amino acid predicted for a receptor which couples to phospholipase-C. Thus, sequence conservation between CaRs and mGluRs appears to be consistent mostly with conservation of structural domains involved in ligand binding and G-protein coupling and does not provide evidence for specific sequence motifs within intracellular regions predictive of G-protein coupling specificity. Cell lines expressing CaRs have been obtained and their use to identify compounds which modulate the activity of CaRs disclosed (pending U.S. patent application Ser. No. 08/353,784, filed Dec. 9, 1994, hereby incorporated by reference herein).
An ideal screening procedure for identifying molecules specifically affecting the activity of different mGluRs would provide cell lines expressing each functional mGluR in such a manner that each was coupled to the same second messenger system and amenable to high throughput screening.
None of the references mentioned herein are admitted to be prior art to the claims.